Transistors & The End of Moore's Law

How does a transistor work? And when will Moore's Law break down?
Supported by TechNYou - check out their great series on the scientific method: bit.ly/19bBX5G
Thanks to A/Prof Morello and UNSW. Find out more here: bit.ly/17wZ7lt

[@gussilva9937]

2023/2024 here. An update on our current transistor size (Thanks Apple M3): 2013: 22nm -> ~50 silicon atoms 2023: 3nm -> ~13 silicon atoms Prof Morello is pretty on point with his prediction with 2025 being ~5 silicon atoms within the space!

[@toby1248]

3nm is just a marketing term. No feature is smaller than 10nm in a so-called 3nm transistor. The improvements have been mainly focused on adding more 3d elements to the transistors so they can continue taking up less space and being more efficient without having to keep making features smaller

[@MAFiA303]

still waiting for the guitar solo

[@PauloConstantino167]

+Saad MAFiA LOL

[@lokatzlikina]

+Saad MAFiA LMAO

[@theduck001]

+Saad MAFiA Pahahahah

[@NikanDragosysSerpenDra]

why do all long haired people are supposed to play guitar.... i only play air guitar. :P

[@zmo1581]

I'd like your comment but it's exactly at 666 and don't want to ruin it

[@YnseSchaap]

Give this man his own channel !

[@Mitche23]

He is a professor at the university, he has a "channel" :)

[@V.D.22]

where is it?

[@Szobiz]

professors can have channels besides their classrooms u know?

[@davidnguyen8579]

the man channel

[@eyeTelevision]

the school of hard rock

[@MauriceChavez353]

This Morello guy is a godsend. Never heard someone explain this so well. Every answer starts on POINT, ends on POINT, gives the full picture and is UNDERSTANDABLE. Kudos

[@gbin21]

What a cool guy.

[@augustuseuropa410]

Totally rad

[@fluoroantimonic9950]

Not verified?!

[@Chris-vr8cd]

When I grow up I want to have hair like him

[@effonstonee]

Can we get an update on this it’s been long enough and this dude is awesome

[@omrisamorali3816]

agreed

[@OmarExplains]

This guy is fucking awesome :D I could listen to him all day!

[@outofthebox9699]

Now a days we have 4nm transistors.

[@محمد-بن-عباس]

What are you doing here omar 😂

[@AuddityHipHop]

I could listen to this guy all day. Wish he was my teacher.

[@P4P5]

Same

[@solid244]

definately

[@dubbadan1]

It really shows the depth of his understanding of the subject, as well as a high level of empathy. True brilliance.

[@anders5611]

Italian Charisma

[@joshualiu8551]

Hes a good lecture, but trust me hes a really bad thesis mentor, cuz hes so busy everyday. He was like, "ow so thats your problem, ok lets look at it next week." Then, next week, hes not even in the country.

[@rammetin99]

this gentleman is really good at explaining, amazing indeed

[@outofthebox9699]

Now a days we have 4nm transistors.

[@otamanlvhs]

He should do more explanations.I could listen to him all day.

[@krzemian]

+otamanlvhs Well, Andrea Morello is your answer :)

[@LoneWolfZ]

I want to go back to the start of school and have this guy explain everything.

[@viruagrawal]

I've studied Physics for over 5 years, worked in a MEMS lab for 1 year and this is the first time I've truly understood transistors.

[@noman1600]

+Viru A it took me 15 minutes while researching Intel's Broadwell.....IDK who taught you, but you obviously never looked.

[@idogtv]

+marshalcraft Good job, you seem to have Viru A completely figured out. I'm amazed at your ability to analyze this guy's life from his comment.

[@Edmocci]

+marshy & eddie wonder why you guys are so threatened by this video? I like this lecturer as well, I've seen him before, he's very good. You obviously don't think so. lol it's jealousy. (It's hard for me to nail emotions that I'm beyond by about 30 years.)

[@marshalcraft]

Ed West Ha lecturer. It's nothing against him. He was merely discussing something. Not explaining anything that shouldn't have already been learened as an undergrad studying mems technologies.

[@daultonbaird6314]

+Viru A Andrea Morello actually told me very little that I had n't already heard numerous times, but His delivery is clear and concise . If he's not making a ton of vids , it's because he is very busy . He is excellent , and that's at least part of the reason that Der selected him.

[@charlesdahmital8095]

Every two years the number of people that say Moore's Law is ending- doubles.

[@stoferb876]

Moore's Second Law??

[@charlesdahmital8095]

@@stoferb876As I currently understand it Moore's second law has more to do with costs doubling and his first law covered quantity doubling.

[@matthewrix1047]

We can call it Dahmitals law

[@charlesdahmital8095]

@@matthewrix1047 Yes, my plan is working. One step closer to my Nobel prize as a Hypothetical Theorist.

[@TranceCore3]

This is the Moor's Law Law

[@MrNabows]

This guys knows a lot, as Feynman said: "You don't truly understand something if you can't explain it simply"

[@cremonster]

This makes me miss college, when you had those few professors that enjoyed teaching and actually inspired you to gain more knowledge. There's nothing worse than a teacher who diminishes the spark in a student to learn

[@aniximander]

Wow, it's a rare skill to explain difficult concepts in such way. It's understandable for people without a technical background and still isn't completely dumbed down. I would have loved to have this guy as a professor!

[@rcdinkins]

This guy gives the best explanation of a gate I have ever heard.

[@practice00]

The professor is absolutely awesome. The way he can explain complex concepts (not necessarily referring to this video) in such simple layman's terms is imho the hallmark of a great mind.

[@natetroxide]

I could listen to this Professor talk all day!

[@Phoenixx-vy7ln]

same

[@HaouasLeDocteur]

More of this guy.

[@NickiRusin]

+HaouasLeDocteur Yes please!

[@GGminigames]

*Moore of this guy ;)

[@shubhamsarkar9680]

THANKS MAN FOR THE LINK.................

[@moomoodeadcow]

@@mlyarka Nice #sauce

[@fitriezwan]

Amen

[@smeechdog]

Love this guy. I'm a frickin' spazz when it comes to this kinda thing...but every time I see this dude I totally get it. Einstein said if you can't explain it to an 8 year old you don't get it. As my scientific understanding is probably that of an 8 year old, I think this chap passes that incredibly high litmus test.

[@agello24]

all these years. i never could understand the purpose of a transistor. now i do.

[@sandarshsrivastav]

+agello24 Was thinking the same thing!

[@Banzay27]

Real tiny lightswitches that make computers go. Now we know.

[@plowed4weeks]

+agello24 look up booline logic. It'll blow your mind. In a good way :)

[@pjamesbda]

+I'M Running Thaangs - found Boolean Logic?...

[@plowed4weeks]

pjamesbda It's what allows you to turn 1's and 0's into logic and start using them for more then just storing numbers.

[@SuperNerd707]

Maybe if he was my high school physics teacher I would have pursued a more science oriented education in university lol. He was great!

[@DrScrubbington]

Maybe that's why I find these scientific KZbin channels to be more educating than actual school

[@VintageLJ]

+InfinityV0rtex It's because you are more interested. In School, you very likely learnt waaaay more than you will for the next 10 years of watching educational videos. You could recall it if required, but not all the time.

[@eddy2561]

+Supernerd7 In High School I was caught up in the the "fumes" .... perfume and gas!! And.......I sucked at math!!

[@TheGodParticle]

I concur

[@DonCDXX]

+Supernerd7 The old idiom goes; "Those who can, do; those who can't, teach." Almost every teacher in high school was someone who went to college and wasn't capable of going further (intellectually and/or financially) and settled for teaching.

[@nannyoggsally]

Best explaination of a transistor I have ever seen or heard.

[@mustafauslu1685]

There was a saying of the brightest scientist : "If you can't explain it simply, you don't understand it well enough." This is actually an experimental lecturer who gets the idea completely, simmer it down and explaining like a mid-school teacher (even though the phenomena is very complex). Kudos to him!

[@Kenneth_H_Olsen]

Morello is the coolest dude in physics ! :)

[@johnfalcon3614]

He has an accent but speaks so fluently. He's a great teacher. Also who else thinks he would make an excellent cast for Dracula?

[@phastings1]

If he auditioned for "Professor Dracula's Tech School Adventures" he would nail it.

[@LootFragg]

Dee question is how to avoid dee garlic altogether to increase dee yield of blood, by quantum tunneling into dee virgin's chambers, leaving dee guards heisenbergly uncertain about my whereabouts, mwuehehe!

[@albripi]

Italian accent, he's italian. That's why I understand him so well :)

[@parthgupte4175]

I am a draaacula blah blah blah blah!

[@Tuberuser187]

People like this guy renew my faith in Humanity.

[@MrDeerings]

Great explanation! We want more from Mr Morello!

[@VeteranVandal]

Well, Professor Morello, I have to admit that from the researcher's I've seen presenting something, you are one of the best ones in doing it. You've explained this in much clearer, concise and overall better way than the professors that were supposed to teach me this (which effectively hasn't happened, because they themselves didn't understand it, if I'm going to be honest).

[@Joel-ee8df]

I could listen to this guy talk for days and I would never get enough of it. Please put him in more of your videos.

[@radiofun232]

Extremely well explained, 5 stars.

[@rogats]

This is amazingly simply explained. Which means the professor IS good!

[@overengineer7691]

I like how this guy explains everything.

[@pikiwiki]

this guy is gold. clear, concise and logical, even with accented English

[@mason4127]

I went through 2 years of moderate level schooling, and this explanation of moore's law is way better than anything they taught in school. If he is not a professor, I demand to work with him.

[@killervampiredoll]

The camera-man seems to be more interested in the man's face than the lecture he's giving :')

[@Alexander5R]

As an EE this was a very clear and simple explanation of how a FET work. Interesting to hear the BJT version. Also interesting that for Intel, the 14 nm node proved very problematic and still launching new products in 2021in that node, while Samsung and TSMC are now making 5 nm chips.

[@krissp8712]

Yeah, I didn't really understand the scale at the time I first watched this, apart from knowing "lower nm is better". But 50 atoms at 22nm is pretty insanely precise already, no wonder he's looking at going into quantum tech instead.

[@Altricksss]

Wait so you're telling me the person in the thumbnail aint the actor who played Loki in Avengers?

[@totty2524]

I thought it was steve vai.

[@nestorlovesguitar]

I wonder why it is always the dumb, "funny" remark that gets the most likes, even in intellectual, serious videos. I wonder what that says about us as a species.

[@zahra-bs2pz]

Néstor Rodríguez who hurt you?

[@agrisimfarming]

@@zahra-bs2pz How is that offensive its just a question

[@CM-xr9oq]

@@agrisimfarming Why does everybody have to be a comedian?

[@roberthopkins1085]

Just by changing the names of the points on a transistor from Base, Collector and Emitter to Source, Drain and Gate has already simplified my understanding of a transistor. Amazing how changing the terminology a little can make all that difference. Thank you greatly. It has taken nearly 10yrs for me to come to an understanding of how this works.

[@nah9585]

Years later, and I agree with the rest, this person is really easy to listen to and even explains technical stuff with technical words without making it sound complicated. Great video. Are there more?

[@norwegcat]

Outstanding! Explained in a way making it very easy to understand.

[@frankhaugen]

He looks like a Norwegian black metal guitar/vocal -ist

[@JackLe1127]

Frank R. Haugen maybe he does that on the side

[@Weisichnich]

Jack Le I really, really do hope he does... Our Transistors are so small electrons tunnel through the wall law of moore will end us all Quantum research or we faaaaaaaaaaall (high pitch screaming here) Judging by that shirt its glam metal though.

[@JackLe1127]

Weisichnich *shreds vigorously

[@frankhaugen]

Weisichnich do you think it's possible to get in touch with him to persuade him to do something like this. I'm pretty sure he's not a metal band member, but it would be cool to have him sing/lipsynch a "lesson" XD

[@Nereus00]

Frank R. Haugen but he is italian.. and norwegians sucks.. they are not even metallers

[@JPxKillz]

I like this guys hair. He explains things very nicely too.

[@ackka]

CS major here - this is the best explanation of Moore's Law I have ever heard.

[@Dahrenhorst]

After more than 60 years of life I think I now finally understood how transistors are working. Thank you very much for this!

[@LazerLord10]

I really don't like how it's called Moore's *law*. There is no force of nature that *forces* us to make transistors smaller and smaller.

[@erictko85]

Or is there??????? Haha just kidding. I agree, its an overly hubristic name that actual physical laws will prove is not a law.

[@ThereIsNoOtherHandleLikeMine]

Scientific "Laws" are predictions made from observations. They are NOT requirements.

[@mobeen360]

Human greed for more power is enough force lol

[@curandas8801]

Many engineers feel this way. It should more accurately be named "Moore's Observation"

[@itsDhiran]

There is human ambition and curiosity is a big enough force...

[@jsnadrian]

Where did you find this guy? He's amazing!

[@SAbbasHaider]

Diameter of a Silicon atom is 0.22 nm . So a 5nm transistor will be approx 26 Silicon atoms thick . So in accordance of Moore's law by 2025 we have to achieve 1nm thickness , which seems quit possible .

[@bramvandenbroeck5060]

I read about 5 websites and still had difficulties to understand what was being said, but i watched this 8 minute video, and i understood everything he said first ime! This man NEEDS his own channel! Professor, if you read this, please consider to make a youtube channel, i think you would accumulate quite an audience!

[@omkarkoti2372]

I've spent a year learning this in my highschool and this guy did it better in 10 mins❤️

[@minibaa]

IBM just announced a 7nm chip today

[@flagman57]

***** Probably not. They wouldn't announce it if it was terrible.

[@MazeFrame]

+MiniBaa Made from Germanium wich is way more expensive

[@Teth47]

+MatzeGamer SiGe chips have been in production for years, that's part of the Hi-K arch.

[@ufotofu9]

+MiniBaa I think it's more of a proof of concept at this point. Don't expect to see them for consumers soon since it took years o go from 22nm to 14nm.

[@noman1600]

+intj123 the chip is not 7NM, the transistor are 7NMs.....I assume that's what the OP meant

[@SAbbasHaider]

2014 - 14nm 2017 - 10nm 2018 - 7nm 2020 - 5nm 2021 - 4nm / 3.5nm So , by 2025 1nm (4 silicon atoms ) seems to be possible in order to follow Moore's law .

[@wyomingptt]

I want this guy to record Hamlet for an audio book.

[@theq4602]

You do know hamlet was set in Denmark?

[@SSmith-fm9kg]

I went through USAF electronics tech school in 1970. The civilian instructor for basic electronics was trying to explain how a transistor worked. It was a very muddled presentation, leaving me terribly confused. I looked at the symbol diagrams on the chalk board for a while, then asked him, "So a transistor is like a switch, turning current on and off, by using an electrical input?" It was almost like he couldn't comprehend my statement. I wish he could have seen this explanation. This is the clearest I've ever seen.

[@democracymanifest3247]

The world needs more people like this man.

[@ToriRocksAmos]

now where is the qubit follow up video!?

[@JMPM55]

Good explanation but why doesn't the camera person zoom out and keep it still so the viewer can see both the instructor and the black (white) board at the same time, instead of panning left and right? I got dizzy for swaying the camera left and right.

[@Nocturnes1984]

I came here to learn what transistors do and here I am, listening to a prof. from Black Sabbath talking about Quantum tunneling. Its 01am.....and...i'm....listening to ..... quantum mechanics... No...I don't understand quantum mechanics but do understand Transistors now.

[@orimorningstar7094]

Airborne dude it's 2:16 AM and I'm listening to my future self

[@jose.anguila]

This teacher is incredible, I can listen to him all day!

[@KrosWyred]

More videos with this man! I learned about Moore's Law and transistors while in college to be an Aircraft Maintenance Technician and would have loved to have this explained so clearly.

3 years after this video is relased... new amd and nvidia gpus will be 14 nm. About %36 smaller than 22 nm.

[@SeanE90]

+Enes Miraç Kaya Because 20nm production was cancelled.

[@wobblysauce]

AMD is working on 12nm atm, and to think basically, a nm is a Silicon atom.

[@iTracti0n]

+wobbly sauce No, a nm is 2 silicon atoms.

[@hey7328]

and intel says 7 nm will be lowest possible. so we only have two years left

[@TheGUARDIANOFFOR]

if you mean in 2 years you get some education NM is actuali still preaty big and we can go for 3d error prediction so nothing happening yet.... Every 2 years people say it cannot get more smaler but it does... Im heard a profesor that sad HDDS whud never hold more then 1GB ever ..... YEA you wont stop progrees... they find the way around or some new technologi or diferent approche ....

[@Xonatron]

How do we get to the qubit video?

[@adfasfuiuiui1056]

12nm chip says "Hi" from 2016.

[@woutersamyn161]

Last

[@opticaldrive2819]

10nm now

[@itsflashtime2001]

IBM announced 5nm transistor a couple days ago.

[@1345-m9s]

holy shit that's trucking amazing

[@klaasbernd]

2019 here!. Be carefull of the marketing terms used 11nm and such. It used to be that it was the international standard to name the next node to the channel length of the transistor. However, this is no longer the case. The result is that if you watch a 11nm chip under the SEM/TEM you will find that the channel length is actually 15nm or more and the increased speed comes from better architecture among other things. Having said that. The have achieved 7nm lines at least (already in the past) (I've seen the pictures) and are indeed busy with going even further. I do not know how far they are with the actual device making. 2years ago there were still some issues with CD uniformity (aka, consistent channel length and switching speed). Source: I work in the industry. Anyway, with the current way the industry works, add 5nm to the channel length. On the other hand. as I've already seen the 7nm lines, that means what moore's law will continue for some 6-8 years.

[@joeymurray1290]

Mario’s explanations are so clear and concise. Brilliant !!

[@vdmjoris]

His explanation on what a transistor is has never been explained to me more clearly than this. I tip my hat to that! But it is his predictions that 7 years later proved to be spot on! Qubits are coming on strong and we are just at the very very early start.

[@amojak]

he describes the operation of an enhancement mode Field Effect Transistor (FET) , a depletion mode works in the opposite way and needs a negative charge on the Gate. The bipolar transistor is a different beast altogether and requires current into the Base , not voltage and has emitter/collector terminals Base = Gate , Source = Emitter , Drain = Collector.

[@Rcvliegen]

+amojak you are completely correct. This is a FET. A bipolar "or normal" transistor works on current, not on voltage difference.

[@maazahmedpoke]

please make a video about how transistors are made to be just a few atoms apart? are quantum mechanics really the biggest problem and not the difficulty (if not impossible ) to make them literally the size of an atom? i hope you respond. i know it's an old upload. it's been bothering me for a long time

[@TehIdiotOne]

+Maaz ahmed We do believe Our engineering is accurate enough to be able to make transistors that small, and he is correct that the problem is once the transistor size gets that small, there's no real way that we know of, to stop Electrons from jumping

[@dallasbarr-hagan6607]

Thanks you just educated me in a massive way.. thank you so much...

[@KLEANTRIX]

I keep coming back to this video after 6 years... I have watched the whole thing like 8 times. Awesome video.

[@thoughte2432]

Now we're at 7nm & 12nm and at around 10 billion transistors per chip (not only on desktop but also on mobile chips)! Some examples (according to the wikipedia): * Apple A13: 8 500 000 000 transistors, 7nm * Qualcomm Snapdragon 8cx / SCX8180: 8 500 000 000, 7nm * AMD Epyc Rome: almost 40 billion! 7nm & 12nm * Nvidia Turing: around 19 000 000 000, 12 nm (i'm not sure about the size)

[@PieceOfPersia]

This guy makes me proud of being italian, even though he kinda enforces the stereotype that we have a really heavy accent lol

[@JohnDemetriou]

You do :) Well In Milan, Torino, Berghamo and Lake Como you do :) That's where I have been and all italians had a strong accent. Also many did not know english, or did not feel comfortable using english even though they knew what they wanted to say :)

[@PieceOfPersia]

Well, that would be true for almost every non-English-speaking country you visit, because it's not easy at all to grasp what a correct accent would sound like, or at least it isn't for a foreigner who hasn't learned English as a child (that's why in Italian schools English is a mandatory subject, and I believe that's true in almost every other European country aswell, but correct me if I'm wrong). Not to mention that there's no such thing as a "correct english accent". "Should I use an American accent or a British accent? Should I go for a southern/ hillbilly kind of accent, or a New Yorker? Should I sound like a Londonese or like a Scottish guy?". You see my point? Most of us just give up and speak as we normally would in our language, in some way that's the "easy route" :P

[@JohnDemetriou]

PieceOfPersia yeah. i did not say your accent ia wrong. its just that you can tell italians greeks, Hispanic and indians by their accent :-) sometimes french as well :-) there is no correct accent. only understandable and gibberish :-)

[@Vatsyayana87]

accents are not a stereotype... either you have them or not and thats based on culture.

[@Vatsyayana87]

anyone that has an accent to me, will hear an accent from me.

[@dharantejav3375]

His Prediction is literally awesome, He said by 2025 we will make a transistor with only 3 or 4 atoms, and now TSMCs has finished the world's first 3nm plant and begins its production by 2022😂. I can't able to imagine how far we'll go in the future with more integration of quantum mechanics. lol

[@BurriedTruth]

He is incredible, discovered him yesterday by watching a Quantum Computing presentation of his.

[@dharantejav3375]

@@BurriedTruth Yeahh, I decided to watch every video of his XD

[@theblueeyedcrooner1576]

Thank you for posting this. Was thinking when he made the prediction…”wonder what they are now?”.

[@sylviam6535]

If 22nm is 50 atoms, 3nm is about 7 atoms.

[@DeTyPeNeOda]

Damn i was shocked by the ending sound effect :S ^^

[@hall87668]

seriously, despite the knowledge gap, this guys ability to simplify for the layman is phenomenal. I learned more in that 8 min video from him then I have from hours of reading!

[@suburbanrapper]

omg..... please do a whole course in electronics... i have tried and tried...and this is the best explanation i have ever seen.... Thank you Jesus..!

[@Farmfield]

Awesome video - but... I thought for sure this would explain why a new Intel i7 4770k (released q2, 2013) isn't (much) faster than my 3770k (q2, 2012) - but sadly no. Is it just me or doesn't it seem like Moore's law, at least in what it should mean for performance, has been at a standstill for quite a while now..?

[@practice00]

The reason for this is because Intel (and to some extent AMD) have started to focus more and more on energy efficiency rather than raw power. And this focus is caused by rise of the mobile market when compared with the desktop market. So new processor families are thought mobile-first and therefore energy efficiency is prime.

[@Farmfield]

practice - That sounds like a plausible explanation. My own thought have been in this direction though I also believe Intel and AMD has hit some kind of barrier with the current architecture seeing the very slow development from the Core2 line of CPU's until today. Personally I find this particularly sucky as I can't upgrade my current 1-1/2 year old workstation without jumping up to a dual Xeon setup with a $2000+ pricetag on CPU's and MB alone. :P

[@DavidBuxton]

Farmfield Also, CPU's today more powerful in multi-tasking. Old CPU's would be good at single tasks but would struggle when multiple tasks were running concurrently. Personally I think we'll see the end of Moore's law in the next 10-20 years. What happens next is anybody's guess.

[@Farmfield]

***** - What's 'old'? I'm looking at the last 3-4 years. From the faster i7 9xx series back in 2010 to the i7 2600k (2011) you get 15% higher performance but from the 2600k to the i7 3770k to the i7 4770k you get about 5-7% per step, even less if you overclock, so that's at best a 15% performance increase over the last 36 months... Basically a standstill...

[@DavidBuxton]

Not disagreeing with you Farmfield. Raw computational power seems to have slowed (despite Moore's law still holding true). Are we witnessing the limit of silicon or are Intel and AMD (as you rightly suggest) moving their focus away from power to mobile computing? Personally, I think the current gen of Intel CPU's are crazy fast, but I still remember having a single core CPU running at 1.2gz.... :)

[@sergiobosi454]

watching this on 7 nm tech from the futureeee

[@PlasmaMongoose]

What I am getting from this video is that in about 10 years or so, quantum laws will prevent transistors from getting any smaller, so we will be needing something new to replace transistors if we ever hope to have a smartphone with a Exabyte capacity.

[@sam2maddy2001]

Yes and Professor Morello and his team are already working on it.

[@drted]

Honestly, Moore's law is probably already over today. It's getting too expensive to scale transistors down further.

[@PlasmaMongoose]

Ted Sanders I know what you mean, my PC is 3 years old, has 8Gb Ram and a 2Tb hard drive, I looked up the latest PCs today and while the Ram has increased, they still only have 2Tb hard drives at the most.

[@zHSVClubby]

PlasmaMongoose What does a HDD have to do with transistor size? Nothing...

[@zHSVClubby]

***** If you read my comment, you will see I was replying to Plasmamongoose, not you. Genius. She was talking about Hard Drives. HDD data is stored on platters, and not on microchips.

[@starcrafter13terran]

I like how this guy speaks clearly and explains things in a way easy to comprehend.

[@LilBudaa]

This guy is probably one of the best teachers in the world. Andrea Morello, You sir, are absolutely awesome.

[@Dave-yl8rf]

I love his accent

[@oishisakana]

What sort of marjor/program do I need to sign up for to just hang around this guy all day and pick up what hes putting down?

[@Penguin_of_Death]

Well...to start with you'd have to move to Australia...

[@dingo9696]

Electrical engineering

[@chuckjls]

Disruptive technology. awesome. Hope I live long enough to see it.

[@alienkishorekumar]

The fact is that, it's already here, but it's not yet here, I mean to say that they are doing it in the Universities.

[@chuckjls]

alienkishorekumar I guess what I meant was I hope I live long enough to buy a computer made from this technology. It'd probably be the size of a cell phone and have the processing speed of a mainframe.

[@PS-Straya_M8]

I would go to university just to listen to this man, very clear explanations!

[@parajared]

Professor Morello is amazing! He describes things in such a way that it just instantly clicks in my head. I love his teaching style.

[@HumbertoRamosCosta]

2019 and we are stucked at 7nm :(

[@AdiiS]

Not bad considering Intel is still stuck at 14nm lol those bozos

[@4.0.4]

TSMC and Samsung are working on 5nm and 3nm. There is experimental stuff on smaller nodes. I think 2021 we'll have 5nm, but they want it by mid-2020.

[@1pcfred]

@@4.0.4 you don't even really have 7 nm now. They're just counting on the fact that you don't have an electron microscope to see it for yourself.

[@Nuclearcx]

Actually there are working 3nm and 5nm designs already. But anyway, Quantum Computers are also really taking off, therefore possibly solving the problem for good if they can be rigged into doing classical computing which _should_ be simple enough.

[@jorgepereira2950]

@@Nuclearcx Go watch the other video of this guy in Veritasium (1). He explains that Quantum Computers will never replace classical computers, their porpuse is to calculate exponentially. That is, to calculate problems that get way to complicated for classical computing duo to the immense size of numbers / variables involved. But that doesn't make it better for everyday use computing.

[@Icathiann]

I wouldn't be able to take him seriously if he was a teacher, I'd expect him to let his hair out of the ponytail and start singing power metal at any second.

[@TheSkipjack95]

I kept thinking the same thing

[@PremiumBlank]

I was reminded of Dr Mobius from Command and Conquer.

[@MrWarlock616]

lol you made my day :D

[@e1123581321345589144]

are you saying that power metal singers make bad teachers? that's cultural discrimination and I won't have any of it.

[@LootFragg]

"So the object becomes fast enough to escape the gravitational puUUOOOAAHHHLLL!!!" **cue squealie**

[@Jackle02]

This video's great, but the camerawork is atrocious. Please fire the cameraman, and just get a wider shot with more information. This video isn't supposed to be eventful or have a lot of action, you don't need so much shaky cam.

[@eddybarker5072]

I wish I would have had the Count teach my Physics classes. He’s really good.

[@lincolnkarim1]

Nicely explained. I wish you'd explain the operation of the bipolar transistor or even the basic junction diode first before getting into the FET transistor. I may help to demonstrate the evolution of the electronic switch. Regardless it was very well presented.

[@jakewoolard1770]

is that Geddy Lee

[@Slarti]

Why not just make chips bigger? That way you can fit more transistors on the chips.

[@TileWonder]

Because it would cost a lot more to produce. The chips would have a higher chance of failure during manufacturing. And because the chips would use a lot more electricity because they're bigger. There's more reasons, but those 3 are the main ones I believe.

[@connorx447]

Nowadays, manufactures now shrink the die node size, i think Intel is down to 22 nm nodes, hence reducing the chip size with every iteration. Sadly tho, the smaller the nodes, the more heat is stored within the chip. So the smaller the node, the more transistors we can fit, but also the more heat it outputs.

[@felixtsang8948]

This is comparing the chips with the rate of moores law. explain about it on 4:25

[@ken0block]

they are making everything small ( thiner iphone, iwatch , google glass , ...) that will make them to work on smaller CPU & chips.

[@luisloyolaboyokan]

the future is the cloudcomputing, you don't need one computer that is ultrafast if you could use a lot of small computers.

[@SebastianLopez-nh1rr]

I feel like watching a Brady's Channel

[@Wumbolo]

Sebastián López I thought it was Brady asking questions

[@Vatsyayana87]

lol this is veritasium

[@subbuktek]

The best explanation of transistors. I come here often. Use it like a reference video.

[@lawrencepacheco3342]

I've been trying to understand transistors all year and he perfectly explained how they work and I understand now

[@CemKalyoncu]

Maybe irrelevant but this is FET (field effect transistor).

[@scowell]

Cem Kalyoncu There are very few bipolar transistors in your computer or phone. Unlike the '60's, '70s, etc today a generic 'transistor' is an IGFET.

[@louistournas120]

scowell I had to look up IGFET (insulated gate field effect transistor). Looks like it is another acronym for MOSFET (metal oxide semiconductor field effect transistor).

[@scowell]

louis tournas Wikipedia covers it nicely... "An insulated-gate field-effect transistor or IGFET is a related term almost synonymous with MOSFET. The term may be more inclusive, since many "MOSFETs" use a gate that is not metal, and a gate insulator that is not oxide. " An alternative would be JFET, not as common today but with its own uses.

[@drmosfet]

I'm sure BJT are at risk as well.

[@outputresistance]

+scowell They are still heavily used in analog circuits, though.

[@Chevifier]

That moment when youre like: haaaaaaaaaa.

[@AppleJacksApple1]

He sounds like a Khajiit from Elder Scrolls

[@Slarti]

I have not found anyone who explains quantum computing or the mechanism of cpu's better than this man.

[@Calastein]

I could literally watch this guy talk forever, fascinating!